The present invention relates to the field of breath alcohol detection to prevent drunk driving, and more particularly to an improved breath alcohol detection system by combining user voice recognition with user breath sampling taken during user given password speech, to insure the unsupervised test measurement based on user voice signature.
Various attempts have been made to develop devises which prevent automobiles and the like from being operated by drunk individuals. A typical breath testing sobriety interlock, of this type is illustrated in U.S. Pat. No. 4,738,333 to collier, teaches a breathalyzer interlock system used for monitoring the performance of an identity confirming the physical act by the operator, said identity confirming act requiring a sufficient degree of skill as to be incapable of performance in fewer than predetermined number of attempts by substantially all persons other then designated operators who possess said degree of skill. Collier physical act identity confirming requires skilled person using at least one of lips, mouth, tongue and vocal chords to generate at least one tone in a specific frequency range. This has proven to be less than completely desirable because it can not surely Identify the operator, since it requires operator of special skill to move a lip or set the mouth at a certain position into a tube to make a certain frequency sound with a vocal cord. The user in some occasion may not be able to produce said frequency, or some one else with little training easily can reproduce said certain vocal frequency.
Another Method and apparatus for sonic breath determination is illustrated in U.S. Pat. No. 5,734,090 to Koppel which teaches a method and a apparatus for verifying an expiratory breath flow utilizing the sonic characteristic of a standardized breath as reference, where a breath sample is analyzed for the presence and absence of predetermined audio frequency components generated by the expiratory process. Breath sample validity is based upon a match between both the required and missing components. This technique is used to determine a certain frequency produced by the human breath exhale to ascertain the air given into the breathalyzer is that of human or bogus air. Which in this case the prior art fails to identify the user.
In addition other methods has been illustrated in U.S. Pat. No. 4,843,377 to Fuller where in a remote location the prisoner under home arrest is being identified by pictorial image taken by a videophone and user voice tone given into a breathalyzer tube, said information is being sent to a monitoring station through standard phone line. Another problem encountered with the system of the ""377 patent is that, it can operate by a skilled person trained to blow into a tube and produce a certain frequency vocal sound into said tube without removal of mouth from the tube. Fuller technique is not easy and practical for use, the production of certain human vocal sound easily can be reproduced by another person. For User Identification The use of video image is a good concept but requires constant human surveillance. U.S. Pat. No. 4,999,613 to Williams which teaches a remote confinement system, unsupervised test to determine the presence, identity or conduct of prisoner, is performed by a transmitter attached to the prisoner and a receiver at the remote location, a voice test performed to identify a specific prisoner and breath alcohol test to determine the prisoner""s soberly. Events which include test results are sent to a monitoring station through standard phone line. Williams et al. System is highly useful and represents a significant improvement in the art, it does not address the problem addressed by the present invention, namely because the monitoring station can not monitor the presence and perform sobriety test to a person to be monitored when said person is in mobile condition or said person is in a automobile. In reference to Williams et al. the prisoner in order to be in compliance must perform test at a predetermined time at a location, the prisoner must inhale and exhale into a tube regardless of being sober or intoxicated. The present invention if user given password does not contain alcohol, it will not be necessary for user to give full lung inhale exhale breath into the voice recognition breathalyzer.
Even though prior art are useful and good inventions, prior arts fail to provide unsupervised test using voice recognition process to determine the identity and breath alcohol content of a vehicle operator during giving spoken password. In addition in prior arts, the monitoring station fails to supervise and give instruction to a mobil vehicle operator to perform soberly test with user voice recognition and fails to identify the driver and locate the vehicle location by use of GPS/GSM.
It is according to the primary objective of the present invention that it provides a voice recognition breathalyzer system which uses the voice of the vehicle""s operator to thereby provide a operator identification and detection of alcohol during the operators given spoken password, if alcohol is present in operators breath during given speech into said breathalyzer, said vehicle ignition will remain inmmobilized. The present invention provides a vehicle ignition interlock CPU which is passive arming.
It is according to present invention additionally the vehicle operator carries a tamper proof wrist mounted RF transceiver with a built-in vibrator to communicate with the vehicle mounted breathalyzer CPU which is able to identify the operator and it""s presence within a particular vehicle and accordingly signal to the operator of a particular vehicle by vibrating or producing audiovisual signal, reminding the operator to produce given voice recognition breath test prior starting the vehicle engine and to signal the particular vehicle operator to perform given verbal breath rolling test during driving.
In a preferred embodiment said voice recognition breathalyzer is in a hand held housing which has a built-in transmitter to communicate with vehicle installed ignition interlock devise to disarm said vehicle ignition interlock and or to start said vehicle engine.
In another embodiment of the invention, said voice recognition breathalyzer is within a mobile phone to communicate with a particular user given breath test data to a monitoring station, or with a vehicle mounted GSM phone unit to disarm said vehicle engine interlock device with DTMF, or voice command, or send RF command signals to said vehicle ignition interlock CPU.
Additionally, if said voice recognition breathalyzer detects alcohol presence during user given password speech, the breathalyzer CPU produce""s audio-visual signals to the operator to give full long inhale exhale breath into said voice recognition breathalyzer, to measure precise breath alcohol content of the operator. If given password breath into said voice recognition breathalyzer does not contain alcohol, it is not necessary for user to give full long inhale exhale breath test, thus eliminating unnecessary deep lunge breath inhale exhale into the breathalyzer unit in order to operate the vehicle.
Finally In the present invention during a rolling test, if vehicle operator fails voice recognition sobriety test, the monitoring station is capable of locating said vehicle and immobilizing said vehicle engine at a safe speed by use off GPS/GSM module. The breathalyzer, the ignition interlock, and GPS/GSM module with antenna, used in the present invention, are of tamper-proof design, unauthorized removal any of the units or tampering with power supply, or harness will immobilize the vehicles engine by interrupting the flow of needed data sent from said units to said vehicle electronic fuel pump, ignition or starter controller CPU unit. In addition in the present invention said vehicle installed GSM unit is capable sending said tamper signal to the monitoring station.
The components used in the present invention, are easily obtainable, for example the Voice recognition processor is commonly used in many different voice operated applications. Breathalyzer sensor are successfully used with many law enforcement agencies applications, and GPS/GSM phones are commonly used today with many communication and monitoring companies and are easily obtainable off the shelf and reasonable priced.
The present invention provides a sobriety test and method which, without requiring human supervision, confirms the identity of a vehicle operator by user voice recognition during given password command, and breath content into a voice recognition breathalyzer. If the password is valid and breath sample is non toxic, the vehicle breathalyzer CPU disables the vehicle ignition interlock and the operator can start the vehicle successfully. If the given speech breath content contains alcohol the operator can not start the vehicle, the voice recognition breathalyzer signals the operator to give a deep long sample of breath to assure a precise reading. If reading result is above a set threshold the operator can not start the vehicle. If the reading result is at a lower then set threshold the operator can start the vehicle The technique avoids the unnecessary deep lunge breath inhale exhale test for every time operator desires to start the vehicle. In addition the invention provides secure operation by identifying the operators voice signature.
In the present invention multiple user voices could be programed into the system. The system could be pre programed into such as it will be not necessary for all user given password to be analyzed for breath alcohol content. The system will be responsive to analyze breath content of specific operators upon their voice recognition by the system. In a preferred embodiment of the present invention, the voice recognition breathalyzer is portable and it""s built in to a remote RF transmitter with a voice recognition circuitry and breath alcohol sensing circuitry, to identify the user and analyze user breath content for alcohol and transmit user test (passed or fail) data signal by user voice recognition and breath sampling. The pass signal is used for disarming a vehicle installed ignition interlock or a car alarm system, or used for starting a vehicle engine. Said vehicle engine immobilizer CPU being passive arming, it will automatically engage the engine interlock circuitry when said vehicle ignition is being turned off. In the present invention, the vehicle Immobilizer CPU passive arming feature could be temporarily bypassed by operators speech command whose voice is not programmed to be tested for breath alcohol.
It is according to the present invention additionally, the vehicle operator carries a tamper proof wrist mounted RF transceiver has a built in vibrator, to communicate with the vehicle mounted breathalyzer CPU to identify the operator, and it""s presence within a particular vehicle and accordingly signal the operator of said particular vehicle by vibrating, or producing audio-visual signal for operator to produce given verbal password into said breathalyzer prior to starting the vehicle engine, or signal the particular vehicle operator to produce given verbal password during rolling breath test.
In addition, in the present invention, the voice recognition breathalyzer could be installed in a mobile phone capable of analyzing user""s breath sampling with user voice recognition for user identification and send said data to a monitoring station, or signaling said vehicle ignition interlock CPU through said vehicle GPS/GSM or pager unit, to disable said vehicle ignition interlock devise, or to start said vehicle engine.
In the present invention, a monitoring station is capable of signaling the user to give a verbal password into said voice recognition breathalyzer mobil phone, during user given verbal speech, if alcohol is detected, said user ID and alcohol detection data through said mobile phone will be sent to a monitoring station.
In a preferred embodiment of the present invention a person to be monitored is carrying a portable video phone, or the vehicle operated by such individual is equipped with a xe2x80x9cmobile video phonexe2x80x9d which will be utilized for additional identification. The video images are sent to a monitoring station through an RF link. If needed, the mobile phone or videophone used in the invention could be connected to medical instruments which are capable of measuring a persons blood pressure, heart rate (EKG) Blood analyses, cholesterol, Blood sugar Etc. and sent said information data to a monitoring station via RF link.
In preferred embodiment of the invention, during a rolling test, if the driver fails to comply or alcohol is detected in the operator""s breath during given password speech said vehicle emergency lights will flash, and the vehicle""s horn will honk. Additionally, the vehicle could be equipped with GPS/GSM or pager unit. The GSM modem sends said vehicle/driver ID and it""s location to a monitoring station. If the driver of the vehicle pulls over and turns the vehicle engine off, the emergency lights stop flashing and the horn stops honking. If the operator does not comply, the monitoring station sends a signal through the GPS/GSM or pager to said vehicle immobilizer CPU to shut down said vehicle ignition interlock at a safe speed, and notify authorities with it""s location.
In the present invention, if person to be tested xe2x80x9cLoses their voicexe2x80x9d or xe2x80x9ccatches a cold,xe2x80x9d their voice may not be recognizable by the system, or if the voice recognition circuitry malfunctions, In this case the operator calls the monitoring station through the vehicle mount GPS/GSM phone to identify themselves ask for system override signal from the monitoring station. And finally in the present invention, the monitoring station is connected to a server and said data information could be accessed by patrol vehicle mount PC or hand held PC through mobile server or by web server.